A Gilbert-like model for the mobile radio channel

Summary form only given, as follows. A novel approach to coding channel synthesis is proposed for the mobile radio channel, under the assumption that the system performance (i.e. P/sub e/ vs. SNR) in the absence of fading is known in advance. As an example the case of CPM with noncoherent demodulation is considered. On the basis of the P/sub e/-vs.-SNR curve, taking into account the threshold effect, two-state coding channel model parameters are obtained. The 'bad' state corresponds to the practical loss of bit synchronization (due to the deep fade), while in the 'good' state the errors are caused by noise and dependent on the SNR. So, a Gilbert-like model for the coding channel is obtained. The model can be used in studying different types of digital speech encoding (DM, CVSDM) for the mobile radio channel.<
>